🔢 mathematics

Split-prime supercongruence at the mixed CM point (1/6, 1/3; 1)

This paper unconditionally proves that for the mixed CM point (1/6, 1/3, 1), the coefficients of the cubed hypergeometric series satisfy a supercongruence modulo $p^4$ for split primes $p \equiv 1 \pmod 3$ , an enhancement beyond the generic Hodge-gap prediction attributed to the $j=0$ CM structure, while also establishing corresponding inert-prime obstructions via modular realizations on $\Gamma_0(3)$ and Cartier identities.

Original authors: Alex Shvets

Published 2026-05-20

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Alex Shvets

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: A Mathematical Magic Trick

Imagine you have a very long, complicated list of numbers (a sequence). In this paper, the author is looking at a specific list generated by a complex mathematical recipe involving a "hypergeometric function" (think of this as a special machine that churns out numbers based on a specific set of rules).

The author's main discovery is a magic trick that happens when you look at this list through the lens of certain prime numbers (like 7, 13, 19, etc.).

The Trick:
If you pick a specific type of prime number (called a "split prime," which are primes that leave a remainder of 1 when divided by 3), and you look at the number in the list at position $N$ , and then compare it to the number at position $N \times p$ (where $p$ is your prime), they are almost identical.

In fact, they are identical up to a very high level of precision. The author proves that the difference between these two numbers is divisible by $p^4$ (p to the fourth power).

Analogy: Imagine two clocks. One shows the time for "Day 1," and the other shows the time for "Day $p$ ." Usually, these times would be wildly different. But for this specific list and these specific primes, the clocks show the exact same time, down to the microsecond, even though $p^4$ is a huge number.

The "Why" and the "How"

The paper is fascinating because this level of precision ( $p^4$ ) is better than expected.

1. The Generic Expectation (The "Standard" Prediction):
Mathematicians have a general rulebook (the Roberts–Rodriguez Villegas framework) that predicts how well these numbers should match. For this type of problem, the rulebook says they should match up to $p^3$ .

Analogy: It's like a mechanic telling you, "This engine should run smoothly for 3,000 miles."

2. The Surprise (The "CM Enhancement"):
This paper proves the numbers match up to $p^4$ . That extra factor of $p$ is a bonus.

Analogy: The mechanic said 3,000 miles, but the engine actually runs perfectly for 30,000 miles.
The Reason: The author traces this "bonus" to a special geometric shape hidden inside the math called an elliptic curve with $j=0$ . This shape has a special symmetry (like a triangle that looks the same when rotated 120 degrees). The author shows that this symmetry forces the numbers to align even more perfectly than the standard rules predicted.

The Two Types of Primes: The Good and The Bad

The paper also explains what happens with the other type of prime numbers (called "inert primes," which leave a remainder of 2 when divided by 3).

The Split Primes (The Good Ones): The magic trick works. The numbers match perfectly ( $p^4$ ).
The Inert Primes (The Bad Ones): The magic trick fails.
- Analogy: Imagine the list is a song. For the "Split Primes," the song repeats perfectly every time you skip ahead. For the "Inert Primes," the song changes key. The author proves that for these primes, the numbers don't just fail to match; they actually swap between two different "versions" of the sequence. If you skip ahead once, you get Version A. If you skip ahead twice, you get Version B. If you skip ahead three times, you are back to Version A.

The Detective Work: How Did He Prove It?

To prove this, the author had to build a bridge between two different worlds of mathematics: Modular Forms (complex functions with deep symmetry) and Number Theory (the study of integers and primes).

Here is the step-by-step journey of the proof, simplified:

The Map (Modular Realization): The author translates the list of numbers into a map of a geometric landscape (specifically, a shape called $\Gamma_0(3)$ ). This allows him to use the tools of geometry to study the numbers.
The Lens (Frobenius Lift): He uses a special mathematical "lens" called the Frobenius operator. This lens helps him zoom in on the behavior of the numbers at the prime level.
The Obstacle (The Stack Point): There is a tricky spot on the map (the point where $j=0$ $j = 0$ ). It's like a crowded intersection where the geometry gets messy.
- The Breakthrough: The author uses the special 3-fold symmetry (the $\mu_3$ -equivariance) of this intersection. He shows that because of this symmetry, the "traffic" (the mathematical residues) is forced to follow a specific pattern. This pattern eliminates the "noise" that usually prevents the numbers from matching perfectly, allowing the extra $p$ factor to appear.
The Bridge (Atkin–Lehner Intertwining): Finally, he uses a mathematical "bridge" to connect two different parts of the landscape. This bridge cancels out the remaining errors, proving that the "magic trick" (the supercongruence) is real and unconditional.

Summary of Claims

Main Claim: For the specific sequence defined in the paper, if you pick a prime $p \ge 7$ where $p \equiv 1 \pmod 3$ , then the $m$ -th term and the $mp$-th term are congruent modulo $p^4$ .
The Bonus: This $p^4$ precision is one step higher than the standard mathematical prediction ( $p^3$ ), caused by the special symmetry of the $j=0$ elliptic curve.
The Counter-Claim: If you pick a prime $p \ge 5$ where $p \equiv 2 \pmod 3$ , this perfect matching is impossible. The sequence instead oscillates between two different branches modulo $p$ .
Method: The proof relies on translating the problem into modular forms, analyzing a specific geometric point with 3-fold symmetry, and using advanced operators (Cartier and Hecke) to show that the errors cancel out perfectly.

The paper does not claim any applications to physics, cryptography, or medicine. It is a pure mathematics result about the hidden patterns in numbers and the geometry that governs them.

Technical Summary: Split-Prime Supercongruence at the Mixed CM Point

Problem Statement
The paper investigates the arithmetic properties of the integer sequence $A^{\text{mix}}_n$ , defined by the coefficient extraction:
$A^{\text{mix}}_n := 108^n [z^n] \, {}_2F_1\left(\frac{1}{6}, \frac{1}{3}; 1; z\right)^3.$
The central problem is to establish supercongruences of the form $A^{\text{mix}}_{mp} \equiv A^{\text{mix}}_m \pmod{p^k}$ for primes $p$ . Specifically, the author focuses on the "mixed" CM point $(1/6, 1/3; 1)$ , which corresponds to the $j=0$ elliptic curve. The generic prediction from the Roberts–Rodriguez Villegas framework for weight-3 hypergeometric motives suggests a congruence exponent of $k=3$ . The paper seeks to determine if the specific CM nature of this point yields a stronger congruence (an "enhancement") and to characterize the behavior of the sequence at inert primes ( $p \equiv 2 \pmod 3$ ).

Methodology
The proof proceeds through a modular realization of the sequence on the modular curve $X_0(3)$ , utilizing the following technical framework:

Modular Realization and Order Drop:
The sequence is realized via the Lagrange–Bürmann inversion formula. The generating function is expressed in terms of modular forms on $\Gamma_0(3)$ . A key structural insight is an "order drop" in the recurrence relation: the third-order Mao–Tian recurrence for the cube of the hypergeometric function factors as $(S-108)L(2)$ , reducing the problem to a second-order recurrence. This allows the sequence to be written as $A^{\text{mix}}_m = [q^m] C_{\text{mix}}(q) H_{\text{mix}}(q)^m$ , where $C_{\text{mix}}$ is a specific linear combination of weight-5 Eisenstein series and $H_{\text{mix}}$ is a modular kernel.
Katz–Dwork Reduction:
The supercongruence is reduced to a scalar congruence involving the Cartier operator $\Lambda_p$ . Using the Frobenius defect $U_p = \log(t(q)^p/t(qp))$ , the problem is transformed into proving three identities:
$\Lambda_p(C_{\text{mix}} U_p^\ell) \equiv 0 \pmod{p^4}, \quad \ell = 1, 2, 3.$
Here, $C_{\text{mix}} = C_0 - 27 u C_0$ , where $C_0$ and $u C_0$ correspond to two distinct Eisenstein lines.
Witt–Cartier Pole Estimate at the Tame Stack Point:
The core technical input is a local analysis at the elliptic point $P_-$ (where $j=0$ and $u=-1/27$ ). The author works on the tame quotient stack $[\text{Spf } \mathbb{Z}_p[\zeta_3][[r]]/\mu_3]$ .
- $\mu_3$ -Equivariance: For split primes ( $p \equiv 1 \pmod 3$ ), the canonical Frobenius lift $\Phi$ is $\mu_3$ -equivariant. This forces the residue index $K$ in the trace-residue formula to satisfy $K \equiv 1 \pmod 3$ .
- Pole Bound: This congruence condition excludes the residues $K=2$ and $K=3$ , forcing the first non-vanishing pole to be at $K \ge 4$ . This "length-three" estimate yields the extra factor of $p$ , upgrading the bound from $p^3$ to $p^4$ .
Atkin–Lehner Intertwining:
To handle the mixed combination $C_{\text{mix}}$ , the author employs a bridge identity derived from the Atkin–Lehner involution $w_3$ . The intertwining relation $T_p w_3^* = \chi_3(p) w_3^* T_p$ allows the cancellation of the two Eisenstein components ( $C_0$ and $uC_0$ ) at split primes ( $\chi_3(p)=1$ ), ensuring the mixed Cartier identities hold unconditionally.

Key Contributions and Results

Main Theorem (Split Primes): For every split prime $p \ge 7$ ( $p \equiv 1 \pmod 3$ ) and every $m \ge 1$ :
$A^{\text{mix}}_{mp} \equiv A^{\text{mix}}_m \pmod{p^4}.$
This result is unconditional. The exponent 4 exceeds the generic weight-3 Hodge-gap prediction of 3. The paper attributes this "CM enhancement" to the $j=0$ elliptic curve and the specific $\mu_3$ -equivariance of the Frobenius lift at the tame stack point.
Inert Prime Obstruction: For inert primes $p \ge 5$ ( $p \equiv 2 \pmod 3$ ), the split-style supercongruence fails.
- Formal Parameter Obstruction: The congruence $\Lambda_p(C_{\text{mix}} H_{\text{mix}}^p) \equiv C_{\text{mix}} H_{\text{mix}} \pmod{p^4}$ cannot hold.
- Coefficient Parity Law: The sequence satisfies a specific parity law modulo $p$ :
  $A^{\text{mix}}_{mp^r} \equiv A^{(\chi_3(p)^r)}_m \pmod p,$
  where $A^{(\varepsilon)}_m$ corresponds to the Eisenstein branch with sign $\varepsilon$ . Specifically, for inert primes, one Frobenius step exchanges the two Eisenstein branches, and two steps return to the original. Consequently, $A^{\text{mix}}_p \equiv 72 \pmod p$ , while $A^{\text{mix}}_1 = 18$ , demonstrating the failure of the congruence even modulo $p$ .
Computational Verification: The paper includes extensive computational checks in SageMath for primes up to 97, verifying the closure of the Cartier identities, the bridge scalar cancellations, and the inert obstruction predictions.

Significance and Claims
The paper claims that for the sequence in this exact form, the $p^4$ supercongruence at split primes is a new result. The significance lies in:

The Exponent: Proving an exponent of 4, which is one higher than the generic weight-3 prediction, and identifying the mechanism (the $j=0$ CM enhancement) responsible for this gain.
The Mechanism: Providing a detailed local stack-level explanation for the extra digit via the $\mu_3$ -equivariant Witt–Cartier pole estimate, which forces the residue index to skip $K=2,3$ .
The Dichotomy: Establishing a sharp contrast between split and inert primes, where the latter exhibit a "branch exchange" behavior rather than a supercongruence.
Methodological Rigor: The proof is unconditional and relies on a synthesis of Katz's theory of canonical subgroups, modular forms on $\Gamma_0(3)$ , and stack-theoretic local analysis, avoiding reliance on unproven conjectures regarding the Dwork congruence for the main theorem.

The author notes that while the first-digit congruence $A^{\text{mix}}_p \equiv 18 \pmod{p^3}$ follows from known results (via Clausen's identity and Beukers' work), the strengthening to $p^4$ for all $m \ge 1$ and the explicit inert-prime obstruction are novel contributions of this work.

The Big Picture: A Mathematical Magic Trick

The "Why" and the "How"

The Two Types of Primes: The Good and The Bad

The Detective Work: How Did He Prove It?

Summary of Claims

More like this